No abstract

Background Administrators and policymakers are increasingly interested in individual placement and support (IPS) as a way of helping people with severe mental illness (SMI) obtain employment or education. It is thus important to investigate the cost-effectiveness to secure that resources are being used properly. Methods In a randomized clinical trial, 720 people diagnosed with SMI were allocated into three groups; (a) IPS, (b) IPS supplemented with cognitive remediation a social skills training (IPSE), and (c) Service as usual (SAU). Health care costs, municipal social care costs, and labor market service costs were extracted from nationwide registers and combined with data on use of IPS services. Cost-utility and cost-effectiveness analyses were conducted with two primary outcomes: quality-adjusted life years (QALY) and hours in employment. Incremental cost-effectiveness ratios (ICER) were computed for both QALY, using participant’s responses to the EQ-5D questionnaire, and for hours in employment. Results Both IPS and IPSE were less costly, and more effective than SAU. Overall, there was a statistically significant cost difference of €9,543 when comparing IPS with SAU and €7,288 when comparing IPSE with SAU. ICER’s did generally not render statistically significant results. However, there was a tendency toward the IPS and IPSE interventions being dominant, that is, cheaper with greater effect in health-related quality of life and hours in employment or education compared to usual care. Conclusion Individual placement support with and without a supplement of cognitive remediation tends to be cost saving and more effective compared to SAU.

NCT01721824.

No abstract

Purpose To investigate the effect of Individual Placement and Support (IPS) according to diagnoses of schizophrenia, bipolar disorder, major depression, substance use disorders, or forensic psychiatric conditions. Methods A systematic search of the literature was conducted in June 2017 and repeated in December 2020. The systematic review included 13 studies. Analyses of pooled original data were based on the six studies providing data (n = 1594). No studies on forensic psychiatric conditions were eligible. Hours and weeks worked were analyzed using linear regression. Employment, and time to employment was analyzed using logistic regression, and cox-regression, respectively. Results The effects on hours and weeks in employment after 18 months were comparable for participants with schizophrenia, and bipolar disorder but only statistically significant for participants with schizophrenia compared to services as usual (SAU) (EMD 109.1 h (95% CI 60.5-157.7), 6.1 weeks (95% CI 3.9-8.4)). The effect was also significant for participants with any drug use disorder (121.2 h (95% CI 23.6-218.7), 6.8 weeks (95% CI 1.8-11.8)). Participants with schizophrenia, bipolar disorder, and any drug use disorder had higher odds of being competitively employed (OR 2.1 (95% CI 1.6-2.7); 2.4 (95% CI 1.3-4.4); 3.0 (95% CI 1.5-5.8)) and returned to work faster than SAU (HR 2.1 (95% CI 1.6-2.6); 1.8 (95% CI 1.1-3.1); 3.0 (95% CI 1.6-5.7)). No statistically significant effects were found regarding depression. Conclusions IPS was effective regarding schizophrenia, bipolar disorder, and substance use disorder; however, the effect on hours, and weeks worked was not statistically significant regarding bipolar disorder. For people with depression the impact of IPS remains inconclusive. Non-significant results may be due to lack of power. Trial Registration: PROSPERO protocol nr. CRD42017060524